INSTRUCTOR WORKBOOK

Magnetic Levitation Experiment for MATLAB /Simulink Users

Standardized for ABET * Evaluation Criteria
Developed by:
Jacob Apkarian, Ph.D., Quanser
Herv Lacheray, M.A.SC., Quanser
Michel Lvis, M.A.SC., Quanser

Quanser educational solutions

are powered by:

Course material
complies with:

Captivate. Motivate. Graduate.

*ABET Inc., is the recognized accreditor for college and university programs in applied science, computing, engineering, and technology; and has provided leadership and quality assurance
in higher education for over 75 years.

PREFACE
Preparing laboratory experiments can be time-consuming. Quanser understands time constraints of teaching
and research professors. Thats why Quansers control laboratory solutions come with proven practical
exercises. The courseware is designed to save you time, give students a solid understanding of various
control concepts and provide maximum value for your investment.
Quanser courseware materials are supplied in two formats:
1. Instructor Workbook provides solutions for the pre-lab assignments and contains typical
experimental results from the laboratory procedure. This version is not intended for the
students.
2. Student Workbook contains pre-lab assignments and in-lab procedures for students.
This courseware is prepared for users of The MathWorkss MATLAB/Simulink software in
conjunction with Quansers QUARC real-time control software. A version of the course
material for National Instruments LabVIEW users is also available.

The courseware for Magnetic Levitation experiment is aligned with the requirements of the
Accreditation Board for Engineering and Technology (ABET), one of the most respected
organizations specializing in accreditation of educational programs in applied science,
computing, science and technology. The Instructor Workbook provides professors with a simple framework
and set of templates to measure and document students achievements of various performance criteria and
their ability to:
Apply knowledge of math, science and engineering
Design and conduct experiments, and analyze and interpret data
Communicate effectively
Use techniques, skills and modern engineering tools necessary for engineering practice

Quanser, Inc. would like to thank Dr. Karl strom from Lund University, Sweden for his immense contribution
to the courseware content and Dr. Hakan Gurocak from the Washington State University Vancouver, for
rewriting the original manual to include embedded outcomes assessment.

The following material provides an abbreviated example of pre-lab assignments and in-lab procedures for
the Magnetic Levitation experiment. Please note that the examples are not complete as they are intended
to give you a brief overview of the structure and content of the course materials you will receive with the
plant.

TABLE OF CONTENTS

PREFACE ...................................................................................................................... PAGE 1

INTRODUCTION TO QUANSER MAGNETIC LEVITATION COURSEWARE SAMPLE ...... PAGE 3
INSTRUCTOR WORKBOOK TABLE OF CONTENTS ....................................................... PAGE 4
BACKGROUND SECTION SAMPLE ............................................................................ PAGE 6
PRE-LAB QUESTIONS SECTION SAMPLE ................................................................... PAGE 7
LAB EXPERIMENTS SECTION SAMPLE ...................................................................... PAGE 8

1. INTRODUCTION TO QUANSER MAGNETIC LEVITATION COURSEWARE SAMPLE

Quanser courseware provides step-by-step pedagogy for a wide range of control challenges. Starting with the
basic principles, students can progress to more advanced applications and cultivate a deep understanding of
control theories. Quanser Magnetic Levitation courseware covers topics, such as:
Modeling the MAGLEV plant from first principles in order to obtain the two open-loop transfer
functions characterizing the system, in the Laplace domain
Linearize the obtained non-linear equation of motion about the quiescent point of operation
Design, through pole placement, a Proportional-plus-Integral (PI) controller for the MAGLEV
electromagnet current in order for it to meet the required design specifications
Design, through pole placement, a Proportional-plus-Integral-plus-Velocity (PIV) controller with feedforward action for the MAGLEV levitated ball position in order for it to meet the required design
specifications
Implement your two controllers in real-time and evaluate their actual performances
Numerically determine the system's actual closed-loop poles, by considering the coil current control
system's dynamics
Every laboratory chapter in the Instructor Workbook is organized into four sections:
Background section provides all the necessary theoretical background for the experiments.
Students should read this section first to prepare for the Pre-Lab questions and for the actual lab
experiments.
Pre-Lab Questions section is not meant to be a comprehensive list of questions to examine
understanding of the entire background material. Rather, it provides targeted questions for
preliminary calculations that need to be done prior to the lab experiments. All or some of the
questions in the Pre-Lab section can be assigned to the students as homework.
Lab Experiments section provides step-by-step instructions to conduct the lab experiments and
to record the collected data.
System Requirements section describes all the details of how to configure the hardware and
software to conduct the experiments. It is assumed that the hardware and software
configuration have been completed by the instructor or the teaching assistant prior to the lab
sessions. However, if the instructor chooses to, the students can also configure the systems by
following the instructions given in this section.
Assessment of ABET outcomes is incorporated into the Instructor Workbook look for indicators such as A-1, A-2
These indicators correspond to specific performance criteria for an outcome. Appendix A of the Instructor
Workbook includes:
- details of the targeted ABET outcomes,
- list of performance criteria for each outcome,
- scoring rubrics and instructions on how to use them in assessment.
The outcomes targeted by the Pre-Lab questions can be assessed using the student work. The outcomes
targeted by the lab experiments can be assessed from the lab reports submitted by the students. These
reports should follow the specific template for content given at the end of each laboratory chapter. This will
provide a basis to assess the outcomes easily.

2.

INSTRUCTOR WORKBOOK TABLE OF CONTENTS

The full Table of Contents of the Quanser Magnetic Levitation Instructor Workbook is shown here:
1. INTRODUCTION
2. MODELING
2.1. BACKGROUND
2.1.1.ELECTRICAL EQUATIONS
2.1.2.NONLINEAR MODEL
2.1.3.LINEAR MODEL
2.2. PRE-LAB QUESTIONS
3. COIL CURRENT CONTROL
3.1. BACKGROUND
3.1.1.SECOND-ORDER RESPONSE
3.1.2.SPECIFICATIONS
3.1.3.COIL CURRENT CONTROL DESIGN
3.1.4.SET-POINT WEIGHTING
3.1.5.INTEGRAL WINDUP
3.2. PRE-LAB QUESTIONS
3.3. LAB EXPERIMENTS
3.3.1.CURRENT CONTROL SIMULATION
3.3.2.CURRENT CONTROL IMPLEMENTATION
3.4. RESULTS
4. BALL POSITION CONTROL
4.1. BACKGROUND
4.1.1.SPECIFICATIONS
4.1.2.BALL POSITION CONTROL DESIGN
4.2. PRE-LAB QUESTIONS
4.3. LAB EXPERIMENTS
4.3.1.BALL POSITION CONTROL SIMULATION
4.3.2.BALL POSITION CONTROL IMPLEMENTATION
4.4. RESULTS
5. SYSTEM REQUIREMENTS
5.1. OVERVIEW OF FILES
5.2. SETUP FOR COIL CURRENT CONTROL SIMULATION
5.3. SETUP FOR IMPLEMENTING COIL CURRENT CONTROL
5.4. SETUP FOR BALL POSITION CONTROL SIMULATION
5.5. SETUP FOR IMPLEMENTING BALL POSITION CONTROL
6. LAB REPORT
6.1. TEMPLATE FOR COIL CURRENT CONTROL REPORT
6.2. TEMPLATE FOR BALL POSITION CONTROL REPORT
6.3. TIPS FOR REPORT FORMAT

7. SCORING SHEETS
7.1. MODELING PRE-LAB QUESTIONS
7.2. CURRENT CONTROL PRE-LAB QUESTIONS
7.3. CURRENT CONTROL LAB REPORT
7.4. BALL POSITION CONTROL PRE-LAB QUESTIONS
7.5. BALL POSITION CONTROL LAB REPORT
APPENDIX A INSTRUCTORS GUIDE
A.1
PRE-LAB QUESTIONS AND LAB EXPERIMENTS
A.1.1. HOW TO USE THE PRE-LAB QUESTIONS
A.1.2. HOW TO USE THE LABORATORY EXPERIMENTS
A.2
ASSESSMENT FOR ABET ACCREDITATION
A.2.1. ASSESSMENT IN YOUR COURSE
A.2.2. HOW TO SCORE THE PRE-LAB QUESTIONS
A.2.3. HOW TO SCORE THE LAB REPORT
A.2.4 ASSESSMENT OF THE OUTCOMES FOR THE COURSE
A.2.5 COURSE SCORE FOR OUTCOME A
A.2.6 COURSE SCORES FOR OUTCOMES B, K AND G
A.2.7 ASSESSMENT WORKBOOK
A.3
RUBRICS
REFERENCES

3. BACKGROUND SECTION - SAMPLE

A schematic of the Magnetic Levitation (MAGLEV) plant is represented in Figure 2.1. As illustrated in Figure
2.1, the positive direction of vertical displacement is downwards, with the origin of the global Cartesian
frame of coordinates on the electromagnet core flat face. Although the ball does have six Degrees Of
Freedom (DOF) in free space, only the vertical, i.e., x-axis, is controlled. It can also be seen that the MAGLEV
consists of two main systems: an electrical and an electro-mechanical.

Figure 2.1: Schematic of the Magnetic Levitation plant.

Electrical Equations
As represented in Figure 2.1, the MAGLEV coil has an inductance Lc and a resistance Rc. Additionally, the
actual system is equipped with a current sense resistor, Rs, that is in series with the coil. The voltage sense,
Vs, is used to measure the current in the coil. The coil current can then be computed using the following
relationship
Using Kirchhoff's voltage law, we obtain the following first-order differential equation
(2.1)

where Rc is the coil resistance, Lc is the coil inductance, Ic is the coil current, vc is the applied coil voltage, and
Rs is the current sense resistance. This can be represented by the first-order transfer function
(2.2)

where Kc is the DC (or steady-state) gain and c is the time constant.

4. PRE-LAB QUESTIONS SECTION - SAMPLE

Coil Current Control
1. A-1, A-2 Find the PI gains for the coil current control, kp,c and ki,c, in terms of n and . Hint: Remember
the standard second order system equation.
Answer 3.1
Outcome
A-1

Solution
The normalized characteristic equation of the closed-loop transfer function in Equation
3.8 is
(Ans. 3.1)
Equating this with the standard second order system, Equation 3.1, gives the
expressions
And

A-2

Solve for kp,c and ki,c to obtain the control gain equations:
(Ans. 3.2)
And
(Ans. 3.3)

2. A-2 Based on the MAGLEV model parameters, Kc and c found in Section 2.2, calculate the control
gains needed to satisfy the time-domain response requirements given in Section 3.1.2.
Answer 3.2
Outcome
A-2

Solution
Substituting the model parameters found in Ans.2.3 and Ans.2.4 and the natural
frequency and damping ratio in Section 3.1.2, into Ans.3.2 and Ans.3.3 generates the
proportional and integral control gains:
(Ans. 3.4)
And
(Ans. 3.5)

5. LAB EXPERIMENTS SECTION - SAMPLE

Ball Position Control Simulation
Experimental Setup
The s_piv_maglev Simulink diagram shown in Figure 4.2 will be used to simulate the closed-loop ball
position control response with the PIV+FF ball position controller and PI coil current control used earlier in
Section 3.1.3. On the actual device, the ball starts when its on the pedestal at a distance Tb. Similarly, in the
simulation the ball begins at Tb. To prevent a sudden jump, the position setpoint initially starts at Tb and
gradually commands a step about the operating air gap. The speed of the step is slowed down by a Rate
Limiter block.

Figure 4.1: Simulink model used to simulate ball position control response.

IMPORTANT: Before you can conduct these experiments, you need to make sure that the lab files are
configured. If they have not been configured already, then go to Section 5 to configure the lab files first.
1. Enter the current control PI gains in Matlab used in Section 5.2 as Kp_c and Ki_c.
2. Enter the feed-forward, proportional, integral and velocity control gains found in Section 4.2 in Matlab
as Kff_b ,Kp_b, Ki_b and Kv_b.
3. Set the Scale Factor Slider Gain to 1.
4. To generate a step reference, go to the Position Setpoint Signal Generator block and set it to the
following:
Signal type = square
Amplitude = 1
Frequency = 0.25 Hz
5. Set the Amplitude (m) gain block to 1e - 3 and the Operating Air Gap Position constant block to -xb0+1e3 to generate a step that goes between 8 and 10 mm (i.e., 1 mm square wave at 0.25 Hz with 9 mm
constant).
6. Open the Ball Position (mm), Coil Current (A), and Coil Voltage (V) scopes.

7. Start the simulation. By default, the simulation runs for 10 seconds. The scopes should be displaying
responses similar to Figure 4.3. Note that in the Ball Position (m) and Coil Current (A) scopes, the yellow
trace is the setpoint (or command) while the purple trace is the simulation.

Figure 4.3: Simulated closed-loop ball position control response

8. B-5, K-2 Generate a Matlab figure showing the Simulated Ball Position response, the current, and the
input voltage.
Data Saving: Similarly as with s maglev pi, after each simulation run each scope automatically saves their
response to a variable in the Matlabrworkspace. The Ball Position (mm) scopes saves its response to the
data_xb variable. The Coil Current (deg) scope saves its response to the variable called data Ic and the
Coil Voltage (V) scope saves its data to the data_Vc variable.
Answer 4.4
Outcome
B-5
K-2

Solution
The simulation was ran correctly if a response similar to Figure Ans. 4.1 was obtained.
The closed-loop position response is shown in Figure Ans.4.1. You can generate this
using the plot_maglev_ball_rsp.m script.

9. K-1, B-9 Measure the steady-state error, the percent overshoot and the peak time of the simulated
response. Does the response satisfy the specifications given in Section 4.1.1? Keep in mind, due to the
Rate Limiter the setpoint is delayed 0.4 seconds. Take that into account. Hint: Use the Matlab ginput
command to take measurements off the figure.

Figure Ans.4.1: Simulated closed-loop ball position control response.

Answer 4.5
Outcome
K-1

Solution
From the response shown in Ans.4.1, it is clear that the steady-state error is zero, thus
ess = 0
Taking measurements from the step that begins at 6.0 seconds, the response
overshoots up to 10.22 mm at 6.47 seconds and then settles down to 10.2 mm (i.e., 2%
of its final value) at 6.49 seconds. Using the equations given in Section 3.1.1, the settling
time is
ts = 6.49 6 = 0.49 s
Using Equation 3.3 with the measurement, we find that the percent overshoot of the
simulated ball position response is

B-9

The simulation settles to 10.2 mm in 0.49 sec given that the step takes 0.4 seconds to
rise form the 6 second mark. Thus the settling time is acceptable. However, the
overshoot goes above the desired percent overshoot listed in Section 4.1.1. Therefore
the response with the PIV+FF controller does not quite match the specifications.

Process control plants for teaching and research

Magnetic Levitation

Coupled Tanks

These plants are ideal for intermediate level teaching. The are also suitable for research
relating to traditional or modern control applications of process control. For more
information please contact info@quanser.com
2013 Quanser Inc. All rights reserved.

INFO@QUANSER.COM

+1-905-940-3575

Solutions for teaching and research. Made in Canada.

QUANSER.COM